An Introduction to the Web Apps

The book’s website, www.pearsonglobaleditions.com/agresti,

links to several new and interactive web-based applets (or
web apps) that run in a browser. These apps are designed
to help students understand a wide range of statistical
concepts and carry out statistical inference. Many of these
apps are featured (often including screenshots) in Activities
throughout the book. The apps allow saving output (such
as graphs or tables) for potential inclusion in homework or
projects.
• The Random Numbers app generates uniform ran-
dom numbers (with or without replacement) from a
user-defined range of integer values and simulates flip- • The various Sampling Distribution apps generate sam-
ping a (potentially biased) coin. pling distributions of the sample proportion or the sam-
ple mean. These apps let users generate samples of
various sizes from a wide range of distributions such as
skewed, uniform, bell-shaped, bimodal, or custom-built.
The apps display the population distribution, the data
distribution of a randomly generated sample, and the
sampling distribution of the sample mean or propor-
tion. With the (repeated) click on a button, one can see
how the sampling distribution builds up one simulated
random sample at a time and, for large sample sizes,
assumes a bell shape. Users can move sliders for sample
size and various population parameters to see the effect
on the sampling distribution. Chapter 7 shows many
screenshots of these apps.
• The Inference for a Proportion and the Inference for a
Mean app carry out statistical inference. They provide
• The Mean vs. Median app allows users to add or delete graphs, confidence intervals and results from z- or t-tests
points from a dot plot as the users explore the effect of for data supplied in summary or original form.
outliers or skew on these two statistics. • The Explore Coverage app uses simulation to demon-
• The Explore Categorical Data and Explore Quantitative strate the concept of the confidence coefficient, both
Data apps provide basic statistics and plots for user-
supplied data.
• The Explore Linear Regression app allows users to
add or delete points from a scatterplot and observe
how the regression line changes for different patterns
or is affected by outliers. The Fit Linear Regression
app allows users to supply their own data, fit a linear
regression model and explore residuals.
• The Guess the Correlation app lets users guess the cor-
relation for a given scatterplot (and find the correlation
between guesses and the true values).
• The Binomial, Normal, t-, Chi-square, and F Distribution
apps visually explore the meaning of parameters for
these distributions. Users can also find probabilities and
percentiles and check them visually on the graph.